A GIS-Based Spatial Evaluation Model for Planning Urban Regeneration Investments

In recent years, the attention towards the use of artificial intelligence techniques, applied across various sectors (from logistics to urban planning, from healthcare to real estate valuation), has highlighted two divergent aspects in the implementation of these tools: on one hand, the advantages that can arise from their intrinsic ability to process big data, in order to enhance efficiency, innovation, and precision, providing strategic benefits to those involved in decision-making processes; on the other hand, the issues related to the possibility of controlling the generated outputs, when it comes to managing and interpreting black boxes that are difficult to be verified. In the field of territorial intervention valuation, the use of “rapid” estimation models, easily replicable even by less experienced users, represents an added value in identifying the best design solutions, especially in public/private partnership operations, so as to define win-win paths for the involved parties. Borrowing from the economics the principles of the urban rent, a procedure for evaluating the temporal evolution of area incidence factors, applied to the city of Rome (Italy), has been proposed. The results, geo-referenced in a GIS environment, provides for an easy-to-consult graphical interface to identify urban areas to be prioritized. The integration of the proposed method with an elastic net regression analysis has allowed the identification of the socio-economic variables that have most influenced the appreciation/depreciation of territorial areas, serving as a useful support for investment decisions (both public and private) and for urban intervention planning choices.